Improving food environments and tackling obesity: A realist systematic review of the policy success of regulatory interventions targeting population nutrition.

BACKGROUND

This systematic review (PROSPERO: CRD42015025276) employs a realist approach to investigate the effect of "real-world" policies targeting different aspects of the food environment that shape individual and collective nutrition.

OBJECTIVES

We were interested in assessing intermediate outcomes along the assumed causal pathway to "policy success", in addition to the final outcome of changed consumption patterns.

DATA SOURCES

We performed a search of 16 databases through October 2015, with no initial restriction by language.

STUDY ELIGIBILITY CRITERIA

We included all publications that reported the effect of statutory provisions aimed at reducing the consumption of energy-dense foods and beverages in the general population. We allowed all methodological approaches that contained some measure of comparison, including studies of implementation progress.

STUDY APPRAISAL AND SYNTHESIS METHODS

We reviewed included studies using the appraisal tools for pre-post and observational studies developed by the National Heart, Lung, and Blood Institute. Given the considerable heterogeneity in interventions assessed, study designs employed, and outcome measures reported, we opted for a narrative synthesis of results.

RESULTS AND IMPLICATIONS

Results drawn from 36 peer-reviewed articles and grey literature reports demonstrated that isolated regulatory interventions can improve intermediate outcomes, but fail to affect consumption at clinically significant levels. The included literature covered six different types of interventions, with 19 studies reporting on calorie posting on chain restaurant menus. The large majority of the identified interventions were conducted in the US. Early results from recent taxation measures were published after the review cut-off date but these suggested more favorable effects on consumption levels. Nevertheless, the evidence assessed in this review suggests that current policies are generally falling short of anticipated health impacts.